Disparity of reported risk of hypersensitivity reactions (HSR) for platinum chemotherapeutic agents

Abstract

e20662 Background: HSR are observed with multiple chemotherapeutic agents and patients should be counseled regarding risk. A wide variation in frequency of platinum HSR is reported in the literature. We sought to confirm the frequency of HSR in the modern era and examine the cause for disparity in reported versus observed risk. Methods: We conducted a retrospective analysis of Quality Variance Reports from 06/2006 to 02/2008 to investigate incidence of HSR to platinum-based therapy at Moores UCSD Infusion Center (IC) and compared data to those historically reported. HSR were graded according to CTCAE v. 3.0. Results: 1,753 doses of platinum agents were administered over this time period. Our IC assumes 0% risk of HSR to cisplatin (0 events in 701 doses), 0.7 % risk to carboplatin (5 events in 689 doses: 2 grade II, 2 grade III, 1 grade V), and 1.4 % risk to oxaliplatin (5 events in 363 doses: 2 grade II, 1 grade III, 2 grade IV). Review of the literature showed a 1 - 20% risk of HSR to cisplatin, provisional to whether it is administered as single agent or in combination. Carboplatin infers a 0.8 - 50% risk of HSR with increased risk after 6 cycles and with longer platinum-free interval. Finally, oxaliplatin bears a 0.5 - 19% risk of HSR with increased risk after 6 infusions. Conclusions: Our own experience is not entirely consistent with the reported frequency of platinum reactions because of the 62-fold disparity of reported percent risk. We found several factors that may explain the disparity: use of emesis prophylaxis may circumvent HSR, authors’ misinterpretation of data or citation errors, under-reporting of adverse events, and differences in defining percent risk of HSR. Some publications document risk of HSR based on the number of patients that react versus documenting risk based on the number of reactions per number of doses. The table below, using our IC data, underscores this point. This data reveals the necessity of standardizing the way in which we report percent risk, since treatment decisions are often based on percent risk. [Table: see text] No significant financial relationships to disclose.